# Towards Understanding and Improving Refusal in Compressed Models via Mechanistic Interpretability

arXiv: https://arxiv.org/abs/2504.04215 · HTML: https://mlsanity.com/paper/2504.04215/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Steering and probing for LLM safety, one of 13 directions in Mechanistic interpretability.
- Builds on 53 of its 76 references that are in the map.
- First posted on arXiv 5 Apr 2025.

## Placed in

- Mechanistic interpretability / [Steering and probing for LLM safety](https://mlsanity.com/mechinterp/steering-and-probing-for-llm-safety/)

## Builds on (most cited first, 30 of 53)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [The Pile: An 800GB Dataset of Diverse Text for Language Modeling](https://mlsanity.com/paper/2101.00027/) (2020)
- [Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding](https://mlsanity.com/paper/1510.00149/) (2015)
- [BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions](https://mlsanity.com/paper/1905.10044/) (2019)
- [Learning both Weights and Connections for Efficient Neural Networks](https://mlsanity.com/paper/1506.02626/) (2015)
- [HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal](https://mlsanity.com/paper/2402.04249/) (2024)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [Representation Engineering: A Top-Down Approach to AI Transparency](https://mlsanity.com/paper/2310.01405/) (2023)
- [Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small](https://mlsanity.com/paper/2211.00593/) (2022)
- [Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!](https://mlsanity.com/paper/2310.03693/) (2023)
- [SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot](https://mlsanity.com/paper/2301.00774/) (2023)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Refusal in Language Models Is Mediated by a Single Direction](https://mlsanity.com/paper/2406.11717/) (2024)
- [Progress measures for grokking via mechanistic interpretability](https://mlsanity.com/paper/2301.05217/) (2023)
- [RWKV: Reinventing RNNs for the Transformer Era](https://mlsanity.com/paper/2305.13048/) (2023)
- [Towards Automated Circuit Discovery for Mechanistic Interpretability](https://mlsanity.com/paper/2304.14997/) (2023)
- [The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets](https://mlsanity.com/paper/2310.06824/) (2023)
- [SmoothLLM: Defending Large Language Models Against Jailbreaking Attacks](https://mlsanity.com/paper/2310.03684/) (2023)
- [Catastrophic Jailbreak of Open-source LLMs via Exploiting Generation](https://mlsanity.com/paper/2310.06987/) (2023)
- [Jamba: A Hybrid Transformer-Mamba Language Model](https://mlsanity.com/paper/2403.19887/) (2024)

## Built on by (most cited first, 7 of 7)

- [From Signal Degradation to Computation Collapse: Uncovering the Two Failure Modes of LLM Quantization](https://mlsanity.com/paper/2604.19884/) (2026)
- [Silent Alarm: A J-Space Protocol for Comparing Danger Recognition Across Models and Quantization Levels](https://mlsanity.com/paper/2607.12792/) (2026)
- [Quality Is Not a Safety Proxy Under Quantization](https://mlsanity.com/paper/2606.10154/) (2026)
- [Which Decisions Low-Bit Quantization Breaks, and How to Predict Them](https://mlsanity.com/paper/2608.06564/) (2026)
- [Mechanistically Interpreting Compression in Vision-Language Models](https://mlsanity.com/paper/2603.25035/) (2026)
- [When Personality Meets Quantization: A Layer-wise MBTI Analysis of Quantized LLMs](https://mlsanity.com/paper/2608.25977/) (2026)
- [Cosine Similarity Is Not Evidence: Measuring the Noise Floor of Interpretability Transfer Under Quantization](https://mlsanity.com/paper/2609.30275/) (2026)

## Abstract (arXiv)

The rapid growth of large language models has spurred significant interest in model compression as a means to enhance their accessibility and practicality. While extensive research has explored model compression through the lens of safety, findings suggest that safety-aligned models often lose elements of trustworthiness post-compression. Simultaneously, the field of mechanistic interpretability has gained traction, with notable discoveries, such as the identification of a single direction in the residual stream mediating refusal behaviors across diverse model architectures. In this work, we investigate the safety of compressed models by examining the mechanisms of refusal, adopting a novel interpretability-driven perspective to evaluate model safety. Furthermore, leveraging insights from our interpretability analysis, we propose a lightweight, computationally efficient method to enhance the safety of compressed models without compromising their performance or utility.